Operational Statistics from Lake Conroe Dam During Harvey Raise Troubling Questions
Two affidavits in a lawsuit filed against the SJRA for flooding downstream residents during Harvey contain statistics that raise several troubling questions about the operation of gates during the storm.
- Did the SJRA wait too long to begin releasing water in significant volumes?
- As a consequence, did it create an unnecessarily high peak discharge?
- Did it maintain high discharge rates longer than it needed?
- As lake levels declined, why did the SJRA continue releasing 2X to 10X more water than it was taking on when it had up to 3 feet of storage capacity in the Lake Conroe?
- Why did it never let the level of Lake Conroe reach its flowage easement max?
- Could different procedures have reduced downstream flooding?
Affidavits of Gilman and Olmos Contain Insights
The first affidavit comes from Chuck Gilman, the SJRA’s Director of Flood Management. It contains a gold mine of statistics. Tables at the end of the affidavit show the date, time, average lake level, total inflow, and total discharge (cubic feet per second), and the exact time of gate changes. The statistics start August 26, 2017 at 10 p.m. They end three days later at the same time. The seven pages of statistics capture a snapshot of the storm and the SJRA’s response hour by hour during Harvey. At the peak, the SJRA recorded changes every 15 minutes.
The second affidavit comes from Hector Olmos, a Principal and Vice President of consulting firm Freese and Nichols, Inc. Olmos helped develop the gate operations policy for at Lake Conroe for the SJRA. The Olmos affidavit contains the same statistical information in Gilman’s. However, it also contains more details of the Gate Operations Policy in place at the time of Harvey. And the two affidavits assert different facts.
Inflow Vs. Outflow and Flowage Easement Max
In the summary that follows, outflow vs. inflow rates are significant. Gilman swore in his affidavit that the gate operation “…policy is programmed so that even in the most extreme situations, peak outflow will never exceed 70% of inflow.” “Peak” is the key word there. Olmos swore in his affidavit that 80% was the limit. However, statistics show that it never significantly exceeded 60%.
As you review the following, keep in mind another key point. SJRA had the ability and authority to increase the lake level to 207, but stopped short at 206.23 for some reason that the affidavits don’t explain.
Summary of Key Statistics and Actions
Key statistics show that:
- Lake Conroe started to rise at 11:30 p.m. on August 26, 2017 in response to 1,722 cfs flowing into the lake.
- After the lake level reached 201.04 feet at 12:15 a.m. on August 27, SJRA first opened its gates at 12:25 a.m. and started releasing 529 cfs.
- After that, inflow generally increased for the next 24.5 hours, though the increases were not a straight line. Inflow fluctuated up and down, likely in response to feeder bands passing over the watershed or variations in readings due to wave and wind action.
- More than 24 hours after the start of the storm, at 1 a.m. on August 28, inflow peaked at 129,065 cfs. By then, the lake level had reached 205.65 feet and the SJRA was releasing 62,082 cfs, less than half of the inflow.
- From that point on, the inflow generally declined, but not in a straight line.
- The water level in Lake Conroe peaked six hours later at 7 a.m., August 28, at 206.23 feet. That’s roughly three-quarters of a foot BELOW the SJRA’s flowage easement.
- After that, water continued to go down for the duration of the storm, but the SJRA continued increasing its release rate for five more hours, until 12 noon on the 28th. The water level was 206.17 feet, almost a foot below its flowage easement. Inflow was 63,986 cfs (less than half the peak), yet discharge peaked at 79,141. So the lake level and inflow were going down, but the discharge rate kept increasing when the lake had room to spare.
- SJRA kept the release rate above 70,000 cfs until 4:15 a.m. the morning of the 29th, more than 16 hours. By then, the lake level had gone down to 204.58. And the discharge rate was still three times higher than the inflow (71,538 cfs discharge vs 20,287 inflow).
- For the rest of the storm, lake level, inflow rates and discharge rates all continued to decline. The table ends at 10 p.m., August 29th. Lake level equaled 203.44, discharge 22,033, and inflow 6,579.
Turning Points in the Storm
During the entire day of August 27th, outflow fluctuated roughly from 16% to 50% of inflow as the inflow kept building relentlessly.
By the morning of the 29th, downstream areas were flooding badly. The SJRA had roughly three feet of extra storage capacity in Lake Conroe within its flowage easement. Yet it kept releasing, on average, 2 – 10X more water than it was taking in. At one point the ratio exceeded 100:1.
Could the SJRA have used more of Lake Conroe’s available storage capacity as lake levels declined to help reduce downstream flooding?
Neither Mr. Gilman’s, nor Mr. Olmos’ affidavits shed light on these issues.
Chuck Gilman inherited this problem. The SJRA did not hire him until well after Harvey.
Also note that conditions during an emergency can be chaotic. Keyboard quarterbacking after the fact is much easier.
If the SJRA wishes to respond to this post, I will print its position verbatim.
Posted by Bob Rehak on 6/11/2020
1017 Days after Hurricane Harvey
The thoughts expressed in this post represent opinions on matters of public concern and safety. They are protected by the First Amendment of the US Constitution and the Anti-SLAPP Statute of the Great State of Texas.